Skip to content

open-craft/xblock-poll

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

XBlock-Poll

A user-friendly way to query students.

This XBlock has been contributed by McKinsey Academy, and is published under the AGPLv3 licence (see the LICENSE file). It has been developed by OpenCraft.

Introduction

This XBlock enables a course author to create survey/poll elements to get feedback from students. The XBlocks can either be poll or survey XBlocks. Poll XBlocks have one question, and a series of answers. Survey XBlocks have several questions and a handful of (terse) answers that a student is expect to answer each one from (Such as 'True', and 'False', or 'Agree' or 'Disagree')

Feature Overview

Survey and Poll are both designed to minimize the amount of fiddling a course author will have to do in order to create the user experience they desire. By default, answers in polls and questions in surveys are able to be enhanced with Markdown (though it is not recommended to do more than line formatting with it) and images. Formatting for images is handled by the XBlock's formatters to keep a consistent and sane user experience.

The feedback section of a poll or survey is shown after a user has completed the block. It, along with a poll block's question field, are intended to make full use of Markdown.

These blocks currently do not support grading.

Installation and configuration

This XBlock relies on Xblock-utils, which should be installed first.

After this, XBlock-poll may be installed using its setup.py, or if you prefer to use pip, running:

$ pip install git+https://github.com/open-craft/xblock-poll.git

You may specify the -e flag if you intend to develop on the repo.

Setting up a course to use Polls and Surveys

To set up a course to use the Poll and Survey XBlocks, first go to your course's outline page in the studio and look for the Advanced settings link in the top menus.

Advanced Settings

Once there, look for the Advanced Modules List and add "poll" and "survey" to it.

Advanced modules configuration

Save your changes, and you may now add a poll by clicking on the Advanced Modules button on the bottom of a unit editing page and selecting either Poll or Survey.

Poll Examples

Polls contain a single question, letting users select from an array of answers for a choice that suits them best.

Below is an example of a poll:

Poll example

When a user makes a selection and hits submit, their choice is sent to the server, and a tally of the results so far is presented to them.

Poll example results

The top choice's percentage is highlighted while the user's selection is marked by a selected (but disabled) radio button on the side.

Poll variations

Polls may have several customizations. Feedback may be added to a poll for display after the user has submitted their answer. Answers may have images associated with them. The question, feedback, and answers are all permitted to contain Markdown (as well as arbitrary HTML) and so may be customized in the studio (as discussed in the next chapter).

Here is an example of a poll that uses markdown in the question and the answers, and which uses images for each of the answers:

Image-only poll

Please note that using only images is not accessible as Poll XBlock does not provide means for specifying alternate text for images. Instead use images and texts:

Image and Label label poll

This poll also contains a feedback section, which is enhanced with Markdown:

Image-only poll results

Or they may have a mix of both.

Mixed label poll

Polls that have a combination of both images and text will autoformat their results so that the bars line up.

Mixed label poll result

Survey Examples

A survey has multiple answers and multiple questions. The same answers are presented for each question.

Below is an example of a survey:

Survey example

When the user hits submit, their answer is sent to the server for tally. The results are then returned to them.

Survey results

The top choice's percentage is shown in orange while the user's selection is marked with a shaded background.

Survey Variations

Surveys, like polls, may have several customizations. As there are multiple questions but the answers are static, questions have the ability to use images, rather than answers. Surveys may also have a Feedback section, just as polls do.

Here is an example of a survey that mixes use of Markdown and images in its questions:

Mixed label survey

...And here's that survey's result page (with a feedback section that uses Markdown):

Mixed label survey results

Using the Studio to create Polls and Surveys

After adding a Poll or Survey to the unit you're editing, click the edit button to bring up the studio menu.

Polls will have a Question field, and both polls and surveys will have a Feedback field.

Poll edit screen

After these fields, there will be a series of line items. Polls will have a set of Answers:

Poll line items

...while Surveys will have both Questions and Answers:

Survey line items

In polls, answers may be given images, while in surveys, questions may be given them. See the examples sections to get an idea of how images look and fit into the layout of a poll or survey.

Questions and answers may be enhanced with Markdown and HTML. This markup is not sanitized, and is treated as trusted input by course authors.

For example, putting this into a Poll's Answer field:

Markdown mixed with HTML

... Would yield this when rendered:

Rendered Markdown mixed with HTML

The following fields are customizable with Markdown and custom HTML on Polls:

  • Question
  • Feedback
  • Answers

The following fields are customizable with Markdown and custom HTML on Surveys:

  • Feedback
  • Questions

You may add another answer (or question, if creating a survey) by clicking the add links at the bottom of the form:

Add links

Questions and answers are listed in the order they're used in the student display. If you need to change the order of questions or answers, do not swap the values of their fields if any students may already have voted, as those votes will be tied to that line item. Instead, use the ordering arrows on the side:

Ordering Arrows

These will allow you to move the question or answer up and down to change its display order.

If you need to remove a question or answer, you can use the Delete link:

Delete Link

Remember: You must have at least two answers (and at least two questions, in the case of polls). If a user has voted on/for the item you've deleted, they will be permitted to vote again, but will not lose progress.

You may also create a poll or survey in which the results are not shown to the user. To do this, click the checkbox for 'Private Results':

Private Results

Notes on Private Results: Users will be able to change their vote on polls and surveys with this option enabled. An analytics event will not be fired upon the student viewing the results, as the results are never visible. A user will see a thank you message, and optionally, any instructor-provided Feedback in an additional "Feedback" section, when they click submit:

Private Results Submission

When you are finished customizing your poll or survey, click the Save button:

Save button

...or to discard your chances, hit Cancel instead:

Cancel button

Assuming you saved, your new poll or survey should be ready to go as soon as you publish your unit's changes. If there are any issues, you will receive an error message specifying anything that may not be quite right yet.

Notes

A poll or survey should not be deployed until its construction is finalized. Changing an answer or question can cause previous respondent's answers to remap and give an inaccurate picture of the responses.

If a poll has changed enough that it leaves a previous voter's choice invalid, their response will be eliminated from the tally upon their next visit, and they will be permitted to vote again. However, they will not lose progress or their score.

Things that could make a poll's previous answers ambiguous include adding or removing a question, or adding or removing an answer.

Analytics

Two events are fired by the XBlocks-- one for viewing the results of a poll, and one for submitting the user's choice.

The resulting events look like this for polls:

{"username": "staff", "host": "precise64", "event_source": "server", "event_type": "xblock.poll.submitted", "context": {"course_user_tags": {}, "user_id": 1, "org_id": "JediAcademy", "module": {"display_name": "Poll"}, "course_id": "JediAcademy/FW301/2015", "path": "/courses/JediAcademy/FW301/2015/xblock/i4x:;_;_JediAcademy;_FW301;_poll;_2d25e451be884aa7a15b33860d7c9647/handler/vote"}, "time": "2015-01-12T19:13:39.199098+00:00", "ip": "10.0.2.2", "event": {"url_name": "2d25e451be884aa7a15b33860d7c9647", "choice": "B"}, "agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10.9; rv:34.0) Gecko/20100101 Firefox/34.0", "page": "x_module"}
{"username": "staff", "host": "precise64", "event_source": "server", "event_type": "xblock.poll.view_results", "context": {"course_user_tags": {}, "user_id": 1, "org_id": "JediAcademy", "module": {"display_name": "Poll"}, "course_id": "JediAcademy/FW301/2015", "path": "/courses/JediAcademy/FW301/2015/xblock/i4x:;_;_JediAcademy;_FW301;_poll;_2d25e451be884aa7a15b33860d7c9647/handler/get_results"}, "time": "2015-01-12T19:13:39.474514+00:00", "ip": "10.0.2.2", "event": {}, "agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10.9; rv:34.0) Gecko/20100101 Firefox/34.0", "page": "x_module"}

... And like this for surveys:

{"username": "staff", "host": "precise64", "event_source": "server", "event_type": "xblock.survey.submitted", "context": {"course_user_tags": {}, "user_id": 1, "org_id": "JediAcademy", "module": {"display_name": "Survey"}, "course_id": "JediAcademy/FW301/2015", "path": "/courses/JediAcademy/FW301/2015/xblock/i4x:;_;_JediAcademy;_FW301;_survey;_e4975240b6c64a1e988bad86ea917070/handler/vote"}, "time": "2015-01-12T19:13:13.115038+00:00", "ip": "10.0.2.2", "event": {"url_name": "e4975240b6c64a1e988bad86ea917070", "choices": {"enjoy": "Y", "learn": "M", "recommend": "N"}}, "agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10.9; rv:34.0) Gecko/20100101 Firefox/34.0", "page": "x_module"}
{"username": "staff", "host": "precise64", "event_source": "server", "event_type": "xblock.survey.view_results", "context": {"course_user_tags": {}, "user_id": 1, "org_id": "JediAcademy", "module": {"display_name": "Survey"}, "course_id": "JediAcademy/FW301/2015", "path": "/courses/JediAcademy/FW301/2015/xblock/i4x:;_;_JediAcademy;_FW301;_survey;_e4975240b6c64a1e988bad86ea917070/handler/get_results"}, "time": "2015-01-12T19:13:13.513909+00:00", "ip": "10.0.2.2", "event": {}, "agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10.9; rv:34.0) Gecko/20100101 Firefox/34.0", "page": "x_module"}

Viewing the Results

When running inside the edX LMS, course staff members have the ability to view results without voting. If you want to grant members of other groups ability to view the results, you can configure the group names in the django settings using the XBLOCK_POLL_EXTRA_VIEW_GROUPS setting, for example:

XBLOCK_POLL_EXTRA_VIEW_GROUPS = ['poll_staff']

Working with Translations

For information about working with translations, see the Internationalization Support section of the Open edX XBlock Tutorial.

Working with Transifex

Prepare your environment:

$ virtualenv -p 3.8 poll-xblock
$ source poll-xblock/bin/activate
$ make requirements

Also ensure that the Transifex client has the proper authentication in the ~/.transifexrc file.

Push new strings to Transifex:

$ make push_translations

To get the latest translations from Transifex:

$ make pull_translations

For testing purposes it's faster to avoid Transifex and work on dummy Esperanto translations:

$ make build_dummy_translations

Running Tests Locally

The Selenium tests in this XBlock requires Firefox 94.0.1 (the exact version can be found in Makefile).

On Linux, it's possible to install it via the command $ make install_linux_dev_firefox, but you'd have to install it on MacOS manually from the Mozilla website, or you can rely on Travis to do that for you on the cloud.

  • Assuming that the correct Firefox version has been installed, you can run the following command for the tests on Linux:
$ make linux_dev_test

or something like that on MacOS:

$ PATH="path/to/firefox/bin/directory:$PATH" make test

To run tests using tox, simply run: tox. Please note that if not all supported python versions are installed, tox will compile and install python from source. For compilation it may need extra dependencies depending on your system.

API for native mobile frontends

Retrieve fixed data for all poll/survey XBlocks in a course:

GET https://<lms_server_url>/api/courses/v1/blocks/?course_id=<course_id>&username=<username>&depth=all&requested_fields=student_view_data

Example poll return value:

"student_view_data": {
    "feedback": "This is feedback message survey.",
    "private_results": false,
    "max_submissions": 1,
    "question": "Did the explanation above make sense to you?",
    "answers": [
        [
            "R",
            {
                "img_alt": "",
                "img": "",
                "label": "Yes, completely"
            }
        ],
        [
            "B",
            {
                "img_alt": "",
                "img": "",
                "label": "Yes, for the most part"
            }
        ],
        [
            "G",
            {
                "img_alt": "",
                "img": "",
                "label": "Not really"
            }
        ],
        [
            "O",
            {
                "img_alt": "",
                "img": "",
                "label": "Not at all"
            }
        ]
    ]
},

Example survey return value:

"student_view_data": {
    "feedback": "This is feedback message survey.",
    "private_results": false,
    "max_submissions": 1,
    "block_name": "Poll2",
    "answers": [
        [
            "Y",
            "Yes"
        ],
        [
            "N",
            "No"
        ],
        [
            "M",
            "Maybe"
        ],
        [
            "1464806559402",
            "Unsure"
        ]
    ],
    "questions": [
        [
            "enjoy",
            {
                "img_alt": "",
                "img": "",
                "label": "Do you think you will use Polls in your course?"
            }
        ],
        [
            "recommend",
            {
                "img_alt": "",
                "img": "",
                "label": "Do you think you will use Surveys in your course?"
            }
        ],
        [
            "learn",
            {
                "img_alt": "",
                "img": "",
                "label": "Do you think the ability to query students is useful?"
            }
        ],
        [
            "1464806513240",
            {
                "img_alt": "",
                "img": "",
                "label": "Do you like taking Polls and/or Surveys?"
            }
        ]
    ]
},

Retrieve current poll tally and current user's vote

GET https://<lms_server_url>/courses/<course_id>/xblock/<poll_xblock_id>/handler/student_view_user_state

Example return value:

{"tally": {"B": 0, "R": 1, "O": 0, "G": 0}, "submissions_count": 1, "choice": "R"}

Retrieve current survey tally and current user's vote

GET https://<lms_server_url>/courses/<course_id>/xblock/<survey_xblock_id>/handler/student_view_user_state

Example return value:

{
  "tally": {
    "enjoy": {
      "Y": 1,
      "M": 0,
      "N": 0
    },
    "learn": {
      "Y": 0,
      "M": 1,
      "N": 0
    },
    "recommend": {
      "Y": 0,
      "M": 0,
      "N": 1
    }
  },
  "submissions_count": 1,
  "choices": {
    "enjoy": "Y",
    "recommend": "N",
    "learn": "M"
  }
}